Determination method and device of diversion point, equipment and storage medium

Abstract

The invention relates to a diversion point determination method and device, equipment and a storage medium, and the method comprises the steps: obtaining a lane relation between all lanes on a road, and determining a node type corresponding to an intersection node of an intersection lane based on the passing direction of the intersection lane under the condition that the lane relation represents the intersection of the lanes; the node type is confluence or divergence; and determining a relative position relationship between the diversion area at the intersection node and the intersection lane, and determining a diversion point corresponding to the diversion area according to the node type for automatic driving under the condition that the node type is verified to be valid according to the relative position relationship. According to the method, after the node type of the intersection node is determined, the node type is further subjected to validity analysis based on the relative position relation between the diversion area and the intersection lane, so that the diversion point is determined under the condition that the node type is valid, the determination accuracy of the diversion point is improved, scene application is not distinguished, and the universality is higher.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and particularly to a method, device, equipment and storage medium for determining a diversion point. Background Art

[0002] In the development process of autonomous driving technology, high-precision maps are indispensable, and positioning and planning need to be carried out with the help of high-precision maps. The recognition and construction of lane lines, lane topological relationships, and lane node types are important components of high-precision map production. In the prior art, when there is an intersection trend between lane lines, according to the intersecting lane lines, it is determined whether the area corresponding to the lane lines is a divergence area or a confluence area. However, the accuracy of the divergence and confluence areas determined only based on the intersecting lane lines is relatively low; or the high-speed divergence and confluence points can also be identified by using the corresponding model of the highway, but this method is only applicable to highways and has low generality. Summary of the Invention

[0003] One of the purposes of the present invention is to provide a method for determining a diversion point to solve the problems of poor accuracy and low generality in determining the diversion point in the prior art. Another purpose is to provide a device and equipment for determining a diversion point.

[0004] In order to achieve the above purposes, the technical solutions adopted by the present invention are as follows:

[0005] A method for determining a diversion point, the method includes: obtaining the lane relationships between each lane on a road, and when the lane relationships represent lane intersections, determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes; the node type is confluence or divergence; determining the relative position relationship between the diversion area and the intersecting lanes at the intersection node, and when verifying that the node type is valid according to the relative position relationship, determining the diversion point corresponding to the diversion area according to the node type for use in autonomous driving.

[0006] According to the above technical means, directly based on the lane relationships and traffic directions of each lane, the node type corresponding to the intersection node of the intersecting lanes is determined, without distinguishing the road scenario, and has higher generality. After determining the node type of the intersection node, further perform an effectiveness analysis on the node type based on the relative position relationship between the diversion area and the intersecting lanes, so as to determine the diversion point when the node type is valid, and the accuracy of determining the diversion point can be improved.

[0007] Further, obtaining the lane relationships between the lanes on the road includes: connecting the ground lane boundary line data in the obtained ground semantic data in the traffic direction to form the lanes; for each of the lanes, dividing them into at least one lane surface according to a preset division method, and determining the lane surface relationships between the lane surfaces in the at least one lane surface corresponding to each lane; determining the center line nodes corresponding to the in-lane driving center lines of the lane surfaces, and determining the node relationships of the center line nodes corresponding to each lane surface according to the lane surface relationships; and determining the lane relationships between the lanes according to the node relationships.

[0008] According to the above technical means, by refining each lane into at least one lane surface, and then based on the lane surface relationships of the at least one lane surface corresponding to each lane, the lane relationships between the lanes are determined, ensuring the accuracy of the determined lane relationships.

[0009] Further, before determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes when the lane relationship represents an intersection, the method further includes: when there is a connection relationship between multiple center line nodes in the node relationship, determining that the lanes corresponding to the lane surfaces where the multiple center line nodes are located are the intersecting lanes, the node where the multiple center line nodes intersect is the intersection node, and the lane relationship is an intersection.

[0010] According to the above technical means, the node relationship corresponding to the lane relationship being an intersection is clarified: there is a connection relationship between multiple center line nodes, that is, the node relationship is a many-to-many relationship. In this way, the intersecting lanes can be accurately determined, improving the efficiency of determining the diversion points.

[0011] Further, determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes includes: determining the shortest distance between the diversion area and the intersection node according to the diversion area data in the ground semantic data; the intersection node is the node among the multiple center line nodes that has a connection relationship with at least two center line nodes; when the shortest distance is less than a preset distance threshold and the traffic direction is the first direction, determining that the node type is divergence; the first direction is the direction of the intersection node towards the at least two center line nodes; when the shortest distance is less than the preset distance threshold and the traffic direction is the second direction, determining that the node type is confluence; the second direction is the opposite direction of the first direction.

[0012] According to the above technical means, it is defined that when the shortest distance between the diversion area and the intersection node is less than the preset distance threshold, the confluence and divergence node is determined. Considering that the significance of manufacturing the confluence and divergence node is relatively small when the distance between the diversion area and the intersection node is too far, the confluence and divergence node is only manufactured when the distance between the diversion area and the intersection node is within the preset distance threshold, which improves the rationality of manufacturing the confluence and divergence node.

[0013] Further, the determination of the relative position relationship between the diversion area and the intersecting lane at the intersection node includes: determining the in-plane driving center line corresponding to the center line node connected to the intersection node as the center line to be analyzed; extending the center line to be analyzed to obtain an extended center line, and when the extended center line is not a preset special type of center line, using a preset relative position algorithm to determine the relative position relationship based on the center line to be analyzed; wherein, the relative position relationship includes: the diversion area is located in the middle of the intersecting lane.

[0014] According to the above technical means, considering that the significance of manufacturing the confluence and divergence node is relatively small when the extended center line is a preset special type of center line, the relative position relationship between the diversion area and the intersecting lane at the intersection node is determined when the extended center line is not a preset special type of center line, which can avoid the manufacturing of the diversion point when the extended center line is a preset special type of center line and improves the rationality of manufacturing the confluence and divergence node.

[0015] Further, the diversion points include a confluence point, a divergence starting point, and a divergence ending point; when verifying that the node type is valid according to the relative position relationship, determining the diversion point corresponding to the diversion area according to the node type includes: determining that the node type is valid when the relative position relationship indicates that the diversion area is located in the middle of the intersecting lane; when the node type is confluence, determining the intersection node as the confluence point; when the node type is divergence, determining the intersection node as the divergence starting point, and using a preset divergence ending point algorithm to determine the divergence ending point based on the control point of the diversion area; the control point of the diversion area is the point on the diversion area that is closest to the intersection node..

[0016] According to the above technical means, only the confluence ending point is manufactured when the node type is confluence, and the divergence starting point and the divergence ending point are manufactured when the node type is divergence, and different diversion points are manufactured according to different node types.

[0017] Further, based on the edge line data of the diversion area, determine the intersection points of the diversion area of the diversion area, and determine the point closest to the intersection point of the diversion area among the edge lines included in the edge line data as the diversion area control point; when the distance between the diversion area control point and the intersection point of the diversion area is less than a preset distance, determine the control node type corresponding to the diversion area control point; based on the control node type, extend the left and right centerlines of the obtained diversion area, and when the extended left and right centerlines intersect at a point, determine the intersection point after the extension intersection as the intersection node, and determine the node type of the intersection node based on the control node type; when the node type is a divergence, use a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point.

[0018] According to the above technical means, by using the edge line data of the diversion area and the left and right centerlines of the diversion area, the confluence and divergence nodes are determined. In this way, it is possible to produce confluence and divergence nodes in the case where there is no lane intersection relationship but there is a diversion area, further supplement the production method of confluence and divergence nodes, and ensure the integrity of the production of confluence and divergence nodes.

[0019] Further, the divergence end point includes a left divergence end point and a right divergence end point. The step of using a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point includes: obtaining the first lane edge line adjacent to the left side of the diversion area and the second lane edge line adjacent to the right side of the diversion area, and respectively determining the first control point and the second control point closest to the diversion area control point and the intersection node on the first lane edge line and the second lane edge line; obtaining the target centerlines on both sides of the diversion area, and determining the target centerline closest to the first control point among the target centerlines as the first target centerline, and the target centerline closest to the second control point as the second target centerline; determining the projection point of the first control point on the first target centerline as the first projection point, and when the distance between the first control point and the first projection point is less than a preset threshold, determining the first projection point as the left divergence end point; determining the projection point of the second control point on the second target centerline as the second projection point, and when the distance between the second control point and the second projection point is less than the preset threshold, determining the second projection point as the right divergence end point.

[0020] According to the above technical means, a specific method for determining the divergence end point is set, the divergence end point is further refined, and the left divergence end point and the right divergence end point are distinguished, so that the determination of the divergence points in the divergence area is more refined.

[0021] Further, the number of side lines included in the side line data is greater than a first value; determining the diversion area intersection points of the diversion area based on the side line data of the diversion area includes: according to the side line data, determining at least one side line in the diversion area with the same side line direction and an azimuth angle less than a preset angle threshold, and selecting a target side line from the at least one side line; when the points where the target side line intersects with side lines in different side line directions satisfy a first preset position condition, determining the intersection points as the diversion area intersection points.

[0022] According to the above technical means, it can effectively avoid determining the diversion area intersection points in the case where the road side lines are blocked or there are errors in recognition, so as to ensure the normal production of the confluence and divergence nodes of the diversion area when the side line data of the diversion area is incomplete.

[0023] Further, the number of side lines included in the side line data is the first value, and determining the diversion area intersection points of the diversion area based on the side line data of the diversion area includes: obtaining the road boundary data of the side line data within a preset range; when the road boundary data indicates the existence of a road boundary, determining the point on the side line included in the side line data that is farthest from the road boundary as the diversion area intersection point.

[0024] Further, after extending the left and right center lines of the obtained diversion area based on the control node type, the method further includes: when the extended left and right center lines do not intersect at a point, obtaining the third target center line closest to the left side of the diversion area and the fourth target center line closest to the right side of the diversion area; determining the projection point of the diversion area control point on the third target center line as the third projection point, and when the third projection point and the third target center line satisfy a second preset position condition, determining the third projection point as the left divergence end point; determining the projection point of the diversion area control point on the fourth target center line as the fourth projection point, and when the fourth projection point and the fourth target center line satisfy the second preset position condition, determining the fourth projection point as the right divergence end point.

[0025] According to the above technical means, if the extended left and right center lines do not intersect at a point, the preset divergence end point algorithm cannot be used to make the divergence end points. Therefore, the left and right divergence end points are made using the diversion area control points.

[0026] Advantages of the present invention:

[0027] (1) Directly based on the lane relationships and traffic directions of each lane, determining the node types corresponding to the intersection nodes of intersecting lanes, without the need to distinguish road scenarios, and having higher generality;

[0028] (2) After determining the node type of the intersection node, further perform validity analysis on the node type based on the relative positional relationship between the diversion area and the intersecting lanes. In the case where the node type is valid, determine the diversion point corresponding to the diversion area, which can improve the accuracy of determining the diversion point;

[0029] (3) Determine the confluence and divergence node through the side line data of the diversion area and the left and right center lines of the diversion area. In this way, in the case where there is no lane intersection relationship but there is a diversion area, the production of the confluence and divergence node can be carried out, further supplementing the production method of the confluence and divergence node to ensure the integrity of the production of the confluence and divergence node. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic flow chart of a method for determining a diversion point provided by an embodiment of the present invention;

[0031] Figure 2 It is a schematic flow chart of an exemplary method for obtaining lane relationships provided by an embodiment of the present invention;

[0032] Figure 3 It is a schematic diagram of an exemplary lane relationship provided by an embodiment of the present invention;

[0033] Figure 4 It is a schematic flow chart of an exemplary method for determining the node direction type provided by an embodiment of the present invention;

[0034] Figure 5 It is a schematic diagram of an exemplary relative positional relationship provided by an embodiment of the present invention;

[0035] Figure 6 It is a schematic flow chart of an exemplary method for determining the relative positional relationship provided by an embodiment of the present invention;

[0036] Figure 7 It is a schematic flow chart of an exemplary method for determining the divergence end point provided by an embodiment of the present invention Figure 1 ;

[0037] Figure 8 It is a schematic flow chart of an exemplary method for determining the divergence end point provided by an embodiment of the present invention Figure 2 ;

[0038] Figure 9 It is a schematic diagram of an exemplary diversion area intersection point provided by an embodiment of the present invention;

[0039] Figure 10 It is a schematic flow chart of an exemplary method for determining the divergence end point provided by an embodiment of the present invention Figure 3 ;

[0040] Figure 11A schematic flow chart for exemplarily determining the intersection points of the diversion area provided by an embodiment of the present invention Figure 1 ;

[0041] Figure 12 A schematic diagram of the result of exemplarily determining divergence and confluence provided by an embodiment of the present invention Figure 1 ;

[0042] Figure 13 A schematic flow chart for exemplarily determining the intersection points of the diversion area provided by an embodiment of the present invention Figure 2 ;

[0043] Figure 14 A schematic diagram of the result of exemplarily determining divergence and confluence provided by an embodiment of the present invention Figure 2 ;

[0044] Figure 15 A schematic flow chart for exemplarily determining the divergence end point provided by an embodiment of the present invention Figure 4 ;

[0045] Figure 16 A schematic flow chart of a method for determining a diversion point provided by an embodiment of the present invention;

[0046] Figure 17 A schematic structural diagram of a device for determining a diversion point provided by an embodiment of the present invention;

[0047] Figure 18 A schematic structural diagram of a device for determining a diversion point provided by an embodiment of the present invention. Detailed implementation manners

[0048] The following will illustrate the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, rather than for limiting the protection scope of the present invention.

[0049] The present invention provides a method for determining a diversion point, which is implemented by a device for determining a diversion point, as Figure 1 shown, and includes the following steps S101 and S102:

[0050] Step S101, obtain the lane relationships between each lane on the road, and when the lane relationship represents that lanes intersect, determine the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes; the node type is confluence or divergence.

[0051] In an embodiment of the present invention, the device for determining the diversion point can obtain the lane relationships between various lanes, and when the lane relationships are intersecting, determine the node type corresponding to the intersection node of the intersecting lanes based on the traffic direction.

[0052] Exemplarily, if the traffic directions are different, the node types corresponding to the intersection nodes of the intersecting lanes are different. Exemplarily, if viewed from the traffic direction, a single-lane road diverges into at least two lanes, the node type is divergence; if viewed from the traffic direction, at least two lanes merge into a single-lane road, the node type is confluence. Of course, it can also be the intersection between multiple lanes, that is, the intersecting lanes include multiple lanes.

[0053] Step S102: Determine the relative position relationship between the diversion area and the intersecting lanes at the intersection node, and when verifying that the node type is valid according to the relative position relationship, determine the diversion point corresponding to the diversion area according to the node type for autonomous driving.

[0054] In an embodiment of the present invention, after the device for determining the diversion point determines the node type of the intersection node, it will further judge the validity of the node type. The specific judgment method is to judge based on the relative position relationship between the diversion area and the intersecting lanes at the intersection node. In this way, the rationality of the diversion point determined by the diversion area can be ensured.

[0055] In an embodiment of the present application, after the device for determining the diversion point determines the diversion point corresponding to the diversion area, when the subsequent vehicle performs autonomous driving, it can effectively and accurately assist autonomous driving, so that the accuracy of autonomous driving is relatively high.

[0056] Compared with the problems of poor accuracy and low generality in determining the diversion point in the prior art, the present invention directly determines the node type corresponding to the intersection node of the intersecting lanes based on the lane relationships and traffic directions of each lane, without distinguishing road scenarios, and has higher generality. After determining the node type of the intersection node, it further performs validity analysis on the node type based on the relative position relationship between the diversion area and the intersecting lanes to determine the diversion point when the node type is valid, which can improve the accuracy of determining the diversion point.

[0057] In some embodiments, when the device for determining the diversion point executes "obtain the lane relationships between various lanes on the road" in the above step S101, as Figure 2 shown, it may include the following steps S201 to S204:

[0058] Step S201: Connect the ground lane edge line data in the obtained ground semantic data in the traffic direction to form each lane.

[0059] In an embodiment of the present invention, the ground semantic data includes ground lane boundary data, and the ground lane boundary data is the solid and dashed line data that constitutes the lane. After obtaining the ground lane boundary data, the determining device of the diversion point will connect the traffic directions to form each lane.

[0060] Step S202: For each lane, divide it into at least one lane surface according to a preset division method, and determine the lane surface relationship between each lane surface in at least one lane surface corresponding to each lane.

[0061] In an embodiment of the present invention, after forming each lane, the determining device of the diversion point will divide each lane into at least one lane surface according to a preset division method. Exemplarily, the determining device of the diversion point can divide each lane according to a certain threshold, and the threshold can be set to 3m, 5m or other values. The length of the divided lane surface is the threshold, and the width is the lane width.

[0062] In an embodiment of the present invention, while dividing each lane into at least one lane surface, the determining device of the diversion point will determine the lane surface relationship between each lane surface in at least one lane surface corresponding to each lane.

[0063] Exemplarily, the lane surface relationship can be the relationship between lane surfaces in the traffic direction for at least one lane surface corresponding to each lane according to a fixed threshold. The lane surface relationship includes at least the following relationships: no relationship, indicating that the distance between two lane surfaces exceeds the fixed threshold and thus there is no relationship; 1 to 1, indicating that two lane surfaces are in the same lane; many to many (M:N), indicating that the number of lanes changes from M to N; 1 to 1.2, indicating that the lane gradually changes. Among them, the fixed threshold can be 1m, 3m or other values, and the specific fixed threshold can be set according to the actual scenario and application requirements, and the present invention does not limit this.

[0064] As Figure 3 shown, lane 31 is divided into at least one lane surface 32-33, and lane 34 is divided into at least one lane surface 35. At this time, the lane surface 32 and the lane surface 33 are in a 1 to 1 relationship, and the lane surface 32 and the lane surface 33 and the lane surface 35 are in a one-to-many relationship, that is, the situation when M is 1 and N is 2.

[0065] Step S203: Determine the center line nodes corresponding to the in-plane driving center lines of each lane surface, and determine the node relationship of the center line nodes corresponding to each lane surface according to the lane surface relationship.

[0066] In an embodiment of the present invention, the determining device of the diversion point will determine the center line nodes corresponding to the in-plane driving center lines in each lane surface. Exemplarily, as Figure 3As shown, the center line nodes corresponding to the in-plane center line 36 of the lane surface 32 are 37 and 38, the center line nodes corresponding to the in-plane center line 39 of the lane surface 33 are 310 and 311, and the center line nodes corresponding to the in-plane center line 312 of the lane surface 35 are 313 and 314.

[0067] In an embodiment of the present invention, the determining device for the diversion point determines the node relationship of the center line nodes corresponding to each lane surface according to the lane surface relationship. Exemplarily, as Figure 3 shown, the lane surface 32 has a one-to-many relationship with the lane surfaces 33 and 35. Then, at this time, there is a connection relationship, that is, a node relationship, between the center line node 38 and the center line nodes 310 and 313.

[0068] Step S204: Determine the lane relationship between each lane according to the node relationship.

[0069] In an embodiment of the present invention, since the node relationship is actually determined based on the ground lane border line data, the determining device for the diversion point can determine the lane relationship between each lane according to the node relationship. In this way, by refining each lane into at least one lane surface, and then determining the lane relationship based on the lane surface relationship between at least one lane surface corresponding to each lane, the accuracy of the lane relationship determination is ensured.

[0070] In some embodiments, before the determining device for the diversion point executes the step of "when the lane relationship represents lane intersection, determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes" in the above step S101, the following steps may also be executed: when there is a connection relationship between multiple center line nodes in the node relationship, determine that the lanes corresponding to the lane surfaces where the multiple center line nodes are located are intersecting lanes, the node where the multiple center line nodes intersect is the intersection node, and the lane relationship is lane intersection.

[0071] In an embodiment of the present invention, if there is a connection relationship between multiple center line nodes in the node relationship, referring to Figure 3 the connection relationship between the center line node 38 and the center line nodes 310 and 313, then it indicates that the lanes corresponding to the lane surfaces where the multiple center line nodes (referring to Figure 3 38, 310, 313 in Figure 3 ) are located (referring to Figure 3 32, 33, 35 in

[0072] In some embodiments, when the determining device for the diversion point executes "determine the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes" in the above step S101, as Figure 4 shown, it may include the following steps S401 to S403:

[0073] Step S401: Determine the shortest distance between the diversion area and the intersection node according to the diversion area data in the ground semantic data; the intersection node is a node among multiple centerline nodes that has a connection relationship with at least two centerline nodes.

[0074] In the embodiments of the present invention, the ground semantic data includes diversion area data with a special attribute additional linear value attribute. Exemplarily, the diversion area is as Figure 5 shown by the inverted triangle in. The shortest distance between the diversion area and the intersection node is as Figure 5 shown by 51 in, which is the distance between the vertex (diversion area control point) 52 of the inverted triangle of the diversion area and the intersection node 53.

[0075] Step S402: When the shortest distance is less than the preset distance threshold and the traffic direction is the first direction, determine that the node type is divergence; the first direction is the direction of the intersection node towards at least two centerline nodes.

[0076] In the embodiments of the present invention, as Figure 5 shown, if the shortest distance 51 is less than the preset distance threshold and the traffic direction is the first direction, determine that the node type is divergence; exemparily, the first direction is the direction of the intersection node 53 towards at least two centerline nodes 54 and 55. Among them, the preset distance threshold can be 20m, 30m, or other values. The specific preset distance threshold can be set according to the actual situation and application requirements, and the present invention does not limit this.

[0077] Step S403: When the shortest distance is less than the preset distance threshold and the traffic direction is the second direction, determine that the node type is confluence; the second direction is the opposite direction of the first direction.

[0078] In the embodiments of the present invention, as Figure 5 shown, if the shortest distance 51 is less than the preset distance threshold and the traffic direction is the second direction, determine that the node type is confluence; exemparily, the second direction is the direction from two centerline nodes 54 and 55 to the intersection node 53, that is, the first direction and the second direction are opposite to each other. Among them, the specific preset distance threshold has been discussed and will not be elaborated here.

[0079] In an embodiment of the present invention, the determining device for the diversion point defines that the confluence and divergence node is determined only when the shortest distance between the diversion area and the intersection node is less than a preset distance threshold. Considering that the significance of manufacturing the confluence and divergence node is relatively small when the distance between the diversion area and the intersection node is too far, the confluence and divergence node is manufactured only when the distance between the diversion area and the intersection node is within the preset distance threshold, which improves the rationality of manufacturing the confluence and divergence node.

[0080] In some embodiments, when the determining device for the diversion point executes the step of "determining the relative position relationship between the diversion area and the intersecting lane at the intersection node" in the above step S102, as Figure 6 shown, it may include the following steps S601 and S602:

[0081] Step S601: Determine the in-plane driving center line corresponding to the center line node connected to the intersection node as the center line to be analyzed.

[0082] In an embodiment of the present invention, the determining device for the diversion point determines the in-plane driving center line corresponding to the center line node connected to the intersection node as the center line to be analyzed. Exemplarily, as Figure 3 shown, the determining device for the diversion point determines the in-plane driving center lines (35 and 39) corresponding to the center line nodes (310 and 313) connected to the intersection node 37 as the center lines to be analyzed (35 and 39).

[0083] Step S602: Extend the center line to be analyzed to obtain an extended center line, and when the extended center line is not a preset special type of center line, use a preset relative position algorithm to determine the relative position relationship based on the center line to be analyzed. Among them, the relative position relationship includes: the diversion area is located in the middle of the intersecting lane.

[0084] In an embodiment of the present invention, the determining device for the diversion point extends the center line to be analyzed to obtain an extended center line. If the extended center line is not a preset special type of center line, use a preset relative position algorithm to determine the relative position relationship based on the center line to be analyzed. Among them, the preset special type of center line may be a special type of center line in the intersection area, indicating that the lane ends, and there is no need to manufacture a diversion point for this diversion area here. If the extended center line is not a preset special type of center line, it is still necessary to further judge the relative position relationship between the diversion area and the intersecting lane.

[0085] Exemplarily, the implementation manner of the device for determining the diversion point to determine the relative position relationship based on the center line to be analyzed by using the preset relative position algorithm is as follows: 1. Determine any center line in the center lines to be analyzed as the first center line (line), and determine the center line different from the first center line as the second center line (other line); extend the first center line. If the intersection node is a divergence node, extend it forward, and if it is a confluence node, extend it backward; the extension is stopped when encountering an intersection, a center line bifurcation, or the sum of lengths is greater than the threshold, and the extended first center line (extension_lines) is obtained. 2. Obtain the left and right lane boundary line set attributes (l_dividers, r_dividers) in the extension_lines, and respectively determine whether there is a boundary line divider (relative position relationship) containing the diversion area in the left and right boundary line sets. If the following conditions are met: a) The distance between the intersection node and the boundary line of the diversion area is less than the specified threshold; b) The other_line center line does not exist between the line and the divider; then record it in the temporary set (l_divisions, r_divisions). 3. Respectively obtain the boundary line closest to the intersection node in the set (l_divisions, r_divisions), and the direction (left, right) of the diversion boundary line relative to the center line as the return result. If the directions of the diversion area boundary lines relative to the center line are one on the left and one on the right, it means that the diversion area is in the middle of the intersecting lanes. In fact, each center line in the center lines to be analyzed will be determined as the first center line to judge which side of the diversion area it is on. Only when there is no other center line between the first center line and the center line of the diversion area boundary line, the boundary line of the first center line will be recorded in the temporary set. That is to say, in fact, there are only the lane boundary lines (including the left boundary line of the diversion area) on the left side of the diversion area and the lane boundary lines (including the right boundary line of the diversion area) on the right side in the temporary set. Furthermore, the closest boundary line obtained in step 3 is the lane boundary line on the left side different from the left boundary line of the diversion area and the lane boundary line on the right side different from the right boundary line of the diversion area.

[0086] In some embodiments, the diversion points include confluence points, divergence starting points, and divergence ending points. When the device for determining the diversion points executes the step "Under the condition that the node type is verified to be valid according to the relative position relationship, determine the diversion point corresponding to the diversion area according to the node type" in the above step S102, as Figure 7 shown, it may further include the following steps S701 to S703:

[0087] Step S701: Determine that the node type is valid when the relative position relationship indicates that the diversion area is in the middle of the intersecting lanes.

[0088] In an embodiment of the present invention, if the relative position indicates that the diversion area is located in the middle of the intersecting lanes, it indicates that the node of the intersection node is valid.

[0089] Step S702: In the case where the node type is confluence, determine the intersection node as a confluence point.

[0090] In an embodiment of the present invention, if the node type is valid and it is in the case of confluence, directly determine the intersection node as a confluence point.

[0091] Step S703: In the case where the node type is divergence, determine the intersection node as a divergence starting point, and use a preset divergence end point algorithm to determine the divergence end point based on the control points of the diversion area; the control points of the diversion area are the points on the diversion area that are closest to the intersection node.

[0092] In an embodiment of the present invention, if the node type is valid and it is in the case of divergence, directly determine the intersection node as a divergence starting point, and use a preset divergence end point algorithm to determine the divergence end point based on the control points of the diversion area. Here, the control points of the diversion area are the points on the diversion area that are closest to the intersection node (see Figure 3 52 in

[0093] In some embodiments, as Figure 8 shown, the device for determining the diversion point may further execute the following steps S801 to S804:

[0094] Step S801: Based on the side line data of the diversion area, determine the diversion area intersection points of the diversion area, and determine the point on the side lines included in the side line data that is closest to the diversion area intersection point as the control point of the diversion area.

[0095] In an embodiment of the present invention, if it is in a variable speed lane scenario (the number of lanes does not change) or in the case of missing lane relationships, the device for determining the diversion point may obtain the side line data of the diversion area, and determine the diversion area intersection points of the diversion area based on the side line data, and determine the point closest to the diversion area intersection point as the control point of the diversion area. Exemplarily, as Figure 9 shown, the side line data includes side line 91 and side line 92, and the point where side line 91 and side line 92 extend and intersect is the diversion area intersection point 93. At this time, determine the point on side line 91 and side line 92 that is closest to the diversion area intersection point 93 as the control point 94 of the diversion area.

[0096] Exemplarily, the implementation manner for the device for determining the diversion point to determine the control points of the diversion area is: obtain the closest point of the intersection point (diversion area intersection point) (pt_intersect) of the diversion area and the grouped side line spatial index (control_str_tree) (side line data) as the control point (control_pt) (control point of the diversion area).

[0097] Step S802: When the distance between the diversion area control point and the intersection point of the diversion area is less than a preset distance, determine the control node type corresponding to the diversion area control point.

[0098] In an embodiment of the present invention, if the distance between the diversion area control point and the intersection point of the diversion area is greater than the preset distance, it means that it is meaningless to make a diversion point here. If the distance between the diversion area control point and the intersection point of the diversion area is less than the preset distance, the control node type corresponding to the diversion area control point will be determined according to the relative position percentage between the diversion area control point and the diversion area boundary line.

[0099] Exemplarily, the implementation method for determining the control node type corresponding to the diversion area control point is: obtain the relative position percentage between the control point and the diversion area boundary line. If the relative position percentage is in the range of [0 - 0.3], it means that the control node type is divergence. If the relative position percentage is in the range of [0.7 - 1], it means that the control node type is confluence.

[0100] Step S803: Based on the control node type, extend the left and right centerlines of the obtained diversion area. When the extended left and right centerlines intersect at a point, determine the intersection point as the intersection node, and determine the node type of the intersection node based on the control node type.

[0101] In an embodiment of the present invention, the device for determining the diversion point will first obtain the left and right centerlines of the diversion area. Exemplarily, the obtaining method is: combine the diversion area boundary line id, and sequentially obtain the centerline sets (l_lanes, r_lanes) (left and right centerlines) of the left and right associated diversion area boundary lines. Perform forward and backward topological extension on the centerlines included in the centerline set to obtain the extended set. If it is divergence, extend forward. If it is confluence, extend backward. If there is an intersection centerline (drivetype == 1) in the set, skip it.

[0102] In an embodiment of the present invention, after extending the left and right centerlines, the node information of the left and right centerlines will be obtained, and the extended set will be traversed to exclude the repeated production situation of the same node type and the same diversion area group attribute. If it has not been produced, and the extended left and right centerlines intersect at a point, determine the intersection point as the intersection node, and determine the node type of the intersection node based on the control node type. The control node type here can be the same as the node type. After determining the node type, the id of the affiliated diversion area group will also be assigned, so that the repeated production situation can be avoided.

[0103] Step S804: When the node type is divergence, use the preset divergence end point algorithm to determine the divergence end point based on the diversion area control point.

[0104] In an embodiment of the present invention, when the node types are different, a preset divergence end point algorithm is used to determine the divergence end point based on the control points in the diversion area.

[0105] In some embodiments, the divergence end point includes a left divergence end point and a right divergence end point. When the device for determining the diversion point executes "using a preset divergence end point algorithm to determine the divergence end point based on the control points in the diversion area", as Figure 10 shown, the following steps S1001 to S1004 may further be included:

[0106] Step S1001: Obtain the first lane boundary adjacent to the left side of the diversion area and the second lane boundary adjacent to the right side of the diversion area, and respectively determine the first control point and the second control point on the first lane boundary and the second lane boundary that are closest to the control points in the diversion area and the intersection node.

[0107] In an embodiment of the present invention, the method for determining the diversion point obtains the first lane boundary adjacent to the left side of the diversion area and the second lane boundary adjacent to the right side of the diversion area, that is, those determined when determining the relative position relationship between the diversion area and the intersecting lane; then, the first control point and the second control point that are closest to the control points in the diversion area and the intersection node are respectively determined on the first lane boundary and the second lane boundary. That is to say, the first control point is on the first lane boundary and the second control point is on the second lane boundary.

[0108] Step S1002: Obtain the target center lines on both sides of the diversion area, and determine the target center line closest to the first control point among the target center lines as the first target center line, and the target center line closest to the second control point as the second target center line.

[0109] In an embodiment of the present invention, the target center lines on both sides of the diversion area are obtained, and the target center line closest to the first control point among the target center lines is determined as the first target center line, and the target center line closest to the second control point is determined as the second target center line.

[0110] Step S1003: Determine the projection point of the first control point on the first target center line as the first projection point, and when the distance between the first control point and the first projection point is less than a preset threshold, determine the first projection point as the left divergence end point.

[0111] In an embodiment of the present invention, the projection point of the first control point on the first target center point is determined as the first projection point, and when the distance between the first control point and the first projection point is less than a preset threshold, the first projection point is determined as the left divergence end point. Among them, the preset threshold may be 10m, 20m or other values. The specific preset threshold can be set according to the actual scenario and application requirements, and this application does not make a limitation in this regard.

[0112] In an embodiment of the present invention, if the distance between the first control point and the first projection point is greater than a preset threshold,

[0113] Step S1004: Determine the projection point of the second control point on the second target center line as the second projection point, and when the distance between the second control point and the second projection point is less than the preset threshold, determine the second projection point as the right divergence end point.

[0114] In an embodiment of the present invention, the projection point of the second control point on the second target center point is determined as the second projection point, and when the distance between the second control point and the second projection point is less than the preset threshold, the second projection point is determined as the right divergence end point. Among them, the specific preset threshold has been discussed and will not be elaborated here.

[0115] Exemplarily, using the preset divergence end point algorithm, the implementation method of determining the divergence end point based on the diversion area control points can be: 1. Obtain the diversion area lane side lines (the first lane side line and the second lane side line) acting on the intersection point; 2. Obtain the starting point coordinates of the diversion area side line as the analysis point (the diversion area control point); 3. Return the point closest to the analysis point (the diversion area control point) among the nodes (the intersection nodes) as the control points (the first control point and the second control point); 4. Obtain the center line sets of the left and right sides of the diversion area (the target center lines), and combine the control points (the first control point and the second control point) to obtain the center lines with the closest left and right distances as the target center lines (the first target center line and the second target center line); 5. Traverse the target center lines. If the distance between the control point and the projection point pt_project (the first projection point and the second projection point) of the target center line is greater than the specified threshold (20m) (the preset threshold), then exclude it. Otherwise, intercept the center line (remove the first and last 10 meters of the target center line) and calculate the position (left, right) of the control point relative to the line segment; 6. Save the relevant attributes such as the coordinates of the pt_project point and the relative position (the left divergence end point and the right divergence end point) to provide data support for subsequent center line cutting.

[0116] In some embodiments, when the number of side lines included in the side line data is greater than the first value, when the diversion point determination device executes the step of "determining the diversion area intersection point of the diversion area based on the side line data of the diversion area" in the above step S801, as Figure 11 shown, it may further include the following step S1101 and step S1102:

[0117] Step S1101: According to the side line data, determine at least one side line in the diversion area with the same side line direction and an azimuth angle less than the preset angle threshold, and select a target side line from the at least one side line.

[0118] In an embodiment of the present invention, if the number of side lines included in the side line data is greater than a first value, at least one side line in the diversion area with the same side line direction and an azimuth angle less than a preset angle threshold can be determined, and the target side line with the longest length is selected from the at least one side line. The first value can be 1; since there are partial missing or discontinuous cases in the side line data of the diversion area included in the side line data, therefore, here, the side line with the longest length is selected from at least one side line with the same side line direction and an azimuth angle less than the preset angle threshold as the target side line to determine the intersection point of the diversion area.

[0119] Exemplarily, as Figure 12 shown, the number of side lines in the figure is greater than 1. At this time, 121 is determined as the target side line, where 1 represents an ordinary node, 10 represents a confluence point, the thick black line represents the diversion area, the thin black line represents the lane side line, the black fence represents the road boundary, and 2 is the projection point of the dotted-solid line connection on the center line.

[0120] Step S1102: When the points where the extended target side line intersects the side lines in different side line directions meet the preset position condition, the intersection points are determined as the intersection points of the diversion area.

[0121] In an embodiment of the present invention, the first preset position condition is that the relative position of the intersection point on the target side line or the side lines in different directions is within a preset interval; exemplarily, the preset interval is (0.1 to 0.9). Of course, the preset interval can also be other intervals, which can be specifically set according to the actual situation and application requirements, and the present invention does not limit this.

[0122] Exemplarily, when the number of side lines is greater than 1, the side lines in the diversion area in the same direction (the azimuth angle calculated by the start and end points of the side line, and the angle is less than a certain threshold (preset angle threshold)) are grouped (excluding the cases of multiple and fragmented side lines in the same direction in the diversion area), and the side line pair with the longest length in the same group (target side line) is retained; the remaining side lines (the target side line and the side lines in different side line directions) are grouped and analyzed pairwise; the intersection point pt_intersect (the intersection point) of the side lines is calculated, and the relative position (percentage) of the intersection point (the intersection point) on the side line in the diversion area is calculated, and it is analyzed and judged whether the position is in this interval (0.1 to 0.9) (the first preset position condition). If it meets the condition, it is the intersection point of the diversion area; if it does not meet the condition, it is skipped.

[0123] In some embodiments, when the number of side lines included in the side line data is the first value, when the diversion point determination device executes the step of "determining the intersection point of the diversion area based on the side line data of the diversion area" in the above step S801, as Figure 13 shown, the following steps S1301 and S1302 may further be included:

[0124] Step S1301: Obtain road boundary data within a preset range of the side line data.

[0125] In an embodiment of the present invention, when the number of side lines included in the side line data is a first value, the determination device for the diversion point obtains road boundary data within a preset range. Herein, the first value may be 1, and the preset range can be set according to actual requirements and application scenarios, and the present invention does not limit this.

[0126] Exemplarily, as Figure 14 shown, the number of side lines in the figure is 1. At this time, 141 is one of the included side lines, where 1 represents an ordinary node, 3 represents a divergence point, 10 represents a confluence point, the thick black line represents the diversion area, the thin black line represents the lane side line, the black fence represents the road boundary, and 2 is the projection point of the solid-dashed line connection on the center line.

[0127] Step S1302: When the road boundary data indicates the existence of a road boundary, determine the point on the side line included in the side line data that is farthest from the road boundary as the diversion area intersection point.

[0128] In an embodiment of the present invention, if there is a representation of a road boundary in the road boundary data, the point on the side line included in the boundary data that is farthest from the road boundary is determined as the diversion area intersection point.

[0129] Exemplarily, when the number of side lines in the diversion area is 1, by searching for the nearest boundary data (road boundary data) around the side line (within a preset range) (judging the height between the nearest points to exclude non-planar result data), if it does not exist, skip it; if it exists, calculate the distances between the start and end points (side lines) of the diversion area and the boundary (road boundary), and retain the farthest point as the intersection point pt_intersect (diversion area intersection point).

[0130] In some embodiments, after the determination device for the diversion point executes "extending the left and right center lines of the obtained diversion area based on the control node type" in step S803, as Figure 15 shown, it may further include the following steps S1501 to S1503:

[0131] Step S1501: When the extended left and right center lines do not intersect at a point, obtain the nearest third target center line on the left side of the diversion area and the nearest fourth target center line on the right side of the diversion area.

[0132] In an embodiment of the present invention, if the extended left and right center lines do not intersect at a point, at this time, directly obtain the nearest third target center line on the left side of the diversion area and the nearest fourth target center line on the right side of the diversion area.

[0133] Step S1502: Determine the projection point of the diversion area control point on the third target center line as the third projection point, and when the third projection point and the third target center line meet the second preset position condition, determine the third projection point as the left branch end point.

[0134] In an embodiment of the present invention, the second preset position condition is that the percentage of the projection point located on the target center line is not equal to 0 or 1, and the distance is greater than a preset value; the preset value here can be 3m, 5m, or any other data, and the specific preset value can be set according to the actual situation and actual needs, and the present invention does not limit this.

[0135] In an embodiment of the present invention, the diversion point determination device determines the projection point of the diversion area control point on the third target center line as the third projection point, and when the third projection point and the third target center line meet the second preset position condition, determines the third projection point as the left branch end point.

[0136] Step S1503: Determine the projection point of the diversion area control point on the fourth target center line as the fourth projection point, and when the fourth projection point and the fourth target center line meet the second preset position condition, determine the fourth projection point as the right branch end point.

[0137] In an embodiment of the present invention, the diversion point determination device determines the projection point of the diversion area control point on the fourth target center line as the fourth projection point, and when the fourth projection point and the fourth target center line meet the second preset position condition, determines the fourth projection point as the left branch end point.

[0138] The present invention provides an exemplary method for determining a diversion point, as Figure 16 shown, mainly including steps S1601 to S1605:

[0139] Step S1601: Introduce ground semantic data and lane surface relationship data.

[0140] Here, the ground semantic data includes solid and dashed line data, and the diversion area boundary line data has a special attribute with an additional linear value attribute. The lane relationship surface data (lane surface relationship) forms a longitudinal connection line (connection in the traffic direction) based on the ground boundary line (ground lane boundary line data), and divides each lane according to a certain threshold (preset division method). There may be breaks due to recognition problems in the middle. If it exceeds the threshold, there is no relationship between the lane surfaces of the same lane. The smallest area quadrilateral (lane surface) is constructed between two adjacent connection lines, and the relationship between the lane surfaces in the traffic direction (lane surface relationship) is calculated based on a fixed threshold (which can be preset according to the actual situation). Among them, the lane surface relationship includes the following types: no relationship, if it exceeds the fixed threshold; one-to-one (1:1); many-to-many (M:N); (1:1.2); (1:1.2) is a special case in many-to-many, indicating lane gradient.

[0141] Step S1602: Generate a lane centerline and lane centerline nodes based on the lane relationship surface.

[0142] Here, traverse the geometry of the lane relationship surface, and generate the in-plane centerline lane and the centerline node relationship table based on the principle of extracting the midpoints of the left and right sides of the lane surface. Among them, the connection of the centerline is combined with the lane surface relationship: in the 1:1 relationship, the geometric points of the centerline are merged, and the centerline nodes are updated; in the many-to-many relationship, nodes are recorded at the relationship, and the distance of the relationship surface (the vertical distance in the traffic direction) is judged. When the distance is less than 3m, the connection relationship is represented by the centerline node connection table; when the distance is greater than 3m, geometric topological connection (bifurcation or merging) is performed based on this node (intersection node), and further judge whether the distance between the diversion area semantic data and this node (intersection node) in the traffic direction is less than the specified threshold (preset distance threshold). If the condition is met, the node type is assigned to the divergence and confluence types with initial values. Otherwise, it is a common node type.

[0143] Step S1603: Verify the rationality and effectiveness of the existing divergence and confluence points, and make divergence end points by combining with the diversion area.

[0144] Here, verify the existing divergence and confluence points in step S1604, and traverse the centerline node types generated in step S1604 as divergence (see Figure 14 in 3), and confluence nodes (see Figure 12 in 10). Combine the node type and the connection table to obtain the set of centerline lists (lanes) to be analyzed for the node (centerlines to be analyzed) and traverse them in turn. Judge whether there is a special type of centerline in the intersection area by extending the front and rear topologies. If there is a special type of centerline (not made within the intersection area range), assign the corresponding node to a common node. Otherwise, combine the diversion area semantics (diversion area) and the centerline to be analyzed through the method method get_scene_statusAnalyze and determine the relative position of the diversion area (see step S602 for the specific calculation method), determine whether the center line exists on both sides of the diversion area edge line, if it meets the conditions, it means that the node type is valid, otherwise it is assigned to the ordinary node type; if the current valid node type is divergent, use method calc_differences_end The method calculates the divergence endpoint (see step S1004 for the specific calculation method).

[0145] Step S1604, creating a divergence and confluence point based on the diversion area group;

[0146] Here, since the divergence and merging initialized in the above step S1602 requires a special lane surface relationship to be produced, the divergence and merging point cannot be produced in the speed change lane scenario (the number of lanes has not changed) or when the lane surface relationship is missing, the divergence and merging point needs to be produced based on the diversion area. Exemplarily, the implementation method of producing divergence and merging points based on the diversion area group can be: 1. Filter and select the diversion area edge line set (edge ​​line data of the diversion area), and group the set based on the diversion area group attribute (groupid). Traverse the diversion area edge line groups in turn, build the spatial index (control_str_tree) of the grouped edge line set, and determine the intersection point pt_intersect (diversion area intersection point) based on the number of grouped edges in two situations: the number is greater than 1 (see the discussion of step S1102); the number is one (see the discussion of step 1202). 2. Get the nearest point between the intersection point (pt_intersect) and the spatial index of the grouped edge line (control_str_tree) as the control point (control_pt) (control point of the diversion area), and exclude the situation where the distance (30m) between the control point and the intersection point (intersection point of the diversion area) is too far. Get the relative position percentage of the control point to the diversion area edge line, and get the initialization node type (divergence: [0~0.3], confluence [0.7~1]). 3. Combined with the diversion area edge line id, get the center line set (l_lanes, r_lanes) (left and right center lines) of the left and right associated diversion area edges in turn, and perform front and back topological extension on the center line set to obtain the set (left and right center lines after extension), divergence (front extension), confluence (back extension); skip when there is an intersection center line (drivetype==1) in the set (left and right center lines after extension). 4. Get the node information of the associated center line, traverse the set to exclude the repeated production of the same node type and the same diversion area group attribute. 5. Determine the centerline set and analyze whether there is a point of intersection based on the node type value extension (divergence (front extension), confluence (back extension)). If there is, return the node, modify the node type value of the node, and assign the diversion area group id to which it belongs. If it is a divergence type, call the method in S1603 calc_differences_endCreate a divergence end point; otherwise, perform the next analysis. 6. Obtain the adjacent points (the third projection point and the fourth projection point) of the centerlines (the third target centerline and the fourth target centerline) that are closest to the left and right of the control point, and respectively obtain the percentage near_pos of the adjacent points to the action centerline, excluding the cases where the percentage is equal to 0 or 1, and the case where the distance value is greater than 3m. 7. Store the adjacent points acting on the left and right centerlines, and assign the associated business attributes (centerline id, diversion area grouping id, direction of the diversion area relative to the adjacent point (left, right, etc.)).

[0147] Step S1605: Save the relevant divergence and confluence points.

[0148] Here, the nodes modified in step S1603 will be updated and stored, and the information newly added in step S1603 and step S1604: the divergence end point and the divergence and confluence point information will be newly stored in a file, which is convenient for making centerline nodes in subsequent steps and provides a basis for interrupting the centerline in the subsequent steps.

[0149] The present invention provides a method for determining a diversion point, which obtains the lane relationships between the lanes on a road, and when the lane relationship represents that the lanes intersect, based on the traffic directions of the intersecting lanes, determines the node type corresponding to the intersection node of the intersecting lanes; the node type is confluence or divergence; determines the relative position relationship between the diversion area and the intersecting lanes at the intersection node, and when the node type is verified to be valid according to the relative position relationship, determines the diversion point corresponding to the diversion area according to the node type for use in autonomous driving. After determining the node type of the intersection node, the present invention further performs an effectiveness analysis on the node type based on the relative position relationship between the diversion area and the intersecting lanes, so as to determine the diversion point when the node type is valid, improving the accuracy of determining the diversion point, and the present invention does not distinguish scene applications, having higher generality.

[0150] An embodiment of the present invention provides a device for determining a diversion point, as Figure 17 shown, including:

[0151] A determination module 1701, configured to obtain the lane relationships between the lanes on a road, and when the lane relationship represents that the lanes intersect, based on the traffic directions of the intersecting lanes, determine the node type corresponding to the intersection node of the intersecting lanes; the node type is confluence or divergence;

[0152] A verification module 1702, configured to determine the relative position relationship between the diversion area and the intersecting lanes at the intersection node, and when the node type is verified to be valid according to the relative position relationship, determine the diversion point corresponding to the diversion area according to the node type for use in autonomous driving.

[0153] In an embodiment of the present invention, the determining module 1701 is further configured to connect the passing directions of the ground lane edge line data in the acquired ground semantic data to form each lane; for each lane, divide it into at least one lane surface according to a preset division method, and determine the lane surface relationship between each lane surface in at least one lane surface corresponding to each lane; determine the center line nodes corresponding to the in-plane driving center lines of each lane surface, and determine the node relationship of the center line nodes corresponding to each lane surface according to the lane surface relationship; according to the node relationship, determine the lane relationship between each lane.

[0154] In an embodiment of the present invention, the determining module 1701 is further configured to, when there is a connection relationship between multiple center line nodes in the node relationship, determine that the lane corresponding to the lane surface where the multiple center line nodes are located is an intersecting lane, the node where the multiple center line nodes intersect is an intersection node, and the lane relationship is a lane intersection.

[0155] In an embodiment of the present invention, the determining module 1701 is further configured to determine the shortest distance between the diversion area and the intersection node according to the diversion area data in the ground semantic data; the intersection node is a node among multiple center line nodes that has a connection relationship with at least two center line nodes; when the shortest distance is less than a preset distance threshold and the passing direction is the first direction, determine that the node type is divergence; the first direction is the direction in which the intersection node faces at least two center line nodes; when the shortest distance is less than a preset distance threshold and the passing direction is the second direction, determine that the node type is confluence; the second direction is the opposite direction of the first direction.

[0156] In an embodiment of the present invention, the verification module 1702 is further configured to determine the in-plane driving center line corresponding to the center line node connected to the intersection node as the center line to be analyzed; extend the center line to be analyzed to obtain an extended center line, and when the extended center line is not a preset special type of center line, use a preset relative position algorithm to determine the relative position relationship based on the center line to be analyzed; wherein, the relative position relationship includes: the diversion area is located in the middle of the intersecting lane.

[0157] In an embodiment of the present invention, the diversion points include confluence points, divergence starting points, and divergence ending points; the verification module 1702 is further configured to, when the relative position relationship indicates that the diversion area is located in the middle of the intersecting lane, determine that the node type is valid; when the node type is confluence, determine the intersection node as the confluence point; when the node type is divergence, determine the intersection node as the divergence starting point, and use a preset divergence ending point algorithm to determine the divergence ending point based on the diversion area control point; the diversion area control point is the point on the diversion area that is closest to the intersection node.

[0158] In an embodiment of the present invention, the determining module 1701 is further configured to determine the diversion area intersection points of the diversion area based on the side line data of the diversion area, and determine the point on the side lines included in the side line data that is closest to the diversion area intersection point as the diversion area control point; in the case where the distance between the diversion area control point and the diversion area intersection point is less than a preset distance, determine the control node type corresponding to the diversion area control point; based on the control node type, extend the left and right centerlines of the obtained diversion area, and in the case where the extended left and right centerlines intersect at a point, determine the intersection point after the extension intersection as the intersection node, and determine the node type of the intersection node based on the control node type; in the case where the node type is a divergence, use a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point.

[0159] In an embodiment of the present invention, the divergence end point includes a left divergence end point and a right divergence end point. The determining module 1701 is further configured to obtain the first lane side line adjacent to the left side of the diversion area and the second lane side line adjacent to the right side of the diversion area, and respectively determine the first control point and the second control point on the first lane side line and the second lane side line that are closest to the diversion area control point and the intersection node; obtain the target centerlines on both sides of the diversion area, and determine the target centerline closest to the first control point among the target centerlines as the first target centerline, and the target centerline closest to the second control point as the second target centerline; determine the projection point of the first control point on the first target centerline as the first projection point, and in the case where the distance between the first control point and the first projection point is less than a preset threshold, determine the first projection point as the left divergence end point; determine the projection point of the second control point on the second target centerline as the second projection point, and in the case where the distance between the second control point and the second projection point is less than a preset threshold, determine the second projection point as the right divergence end point.

[0160] In an embodiment of the present invention, the number of side lines included in the side line data is greater than a first value; the determining module 1701 is further configured to determine at least one side line in the diversion area with the same side line direction and an azimuth angle less than a preset angle threshold according to the side line data, and select a target side line from the at least one side line; in the case where the points where the extended target side line intersects the side lines in different side line directions satisfy a first preset position condition, determine the intersection points as the diversion area intersection points.

[0161] In an embodiment of the present invention, the number of side lines included in the side line data is the first value. The determining module 1701 is further configured to obtain the road boundary data of the side line data within a preset range; in the case where the road boundary data indicates the existence of a road boundary, determine the point on the side line included in the side line data that is farthest from the road boundary as the diversion area intersection point.

[0162] In an embodiment of the present invention, the determining module 1701 is further configured to, when the extended left and right centerlines do not intersect at a point, obtain a third target centerline closest to the left side of the diversion area and a fourth target centerline closest to the right side of the diversion area; determine the projection point of the diversion area control point on the third target centerline as the third projection point, and when the third projection point and the third target centerline meet the second preset position condition, determine the third projection point as the left divergence end point; determine the projection point of the diversion area control point on the fourth target centerline as the fourth projection point, and when the fourth projection point and the fourth target centerline meet the second preset position condition, determine the fourth projection point as the right divergence end point.

[0163] An embodiment of the present invention provides an apparatus for determining a diversion point, as Figure 18 shown, the control node includes: a processor 1801, a memory 1802, and a communication bus 1803;

[0164] The communication bus 1803 is used to implement a communication connection between the processor 1801 and the memory 1802;

[0165] The processor 1801 is configured to execute a computer program stored in the memory 1802 to implement the above-mentioned method for determining a diversion point.

[0166] An embodiment of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores one or more computer programs, and the one or more computer programs can be executed by one or more processors to implement the above-mentioned method for determining a diversion point. The computer-readable storage medium may be a volatile memory, such as a random access memory (RAM); or a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); it may also be a respective device including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc.

[0167] Those skilled in the art should understand that the embodiments of the present invention may be provided as a method, a system, or a computer program product. Therefore, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories and optical memories, etc.) containing computer-usable program code.

[0168] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0169] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0170] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0171] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for determining a diversion point, characterized in that, The method includes: Obtaining the lane relationships between each lane on the road, and when the lane relationships represent lane intersections, determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes; the node type is confluence or divergence; Determining the relative position relationship between the guide area and the intersecting lanes at the intersection node, and when verifying that the node type is valid according to the relative position relationship, determining the guide points corresponding to the guide area according to the node type for autonomous driving to use.

2. The method according to claim 1, characterized in that, The obtaining the lane relationships between each lane on the road includes: Connecting the traffic directions of the ground lane edge line data in the obtained ground semantic data to form each lane; For each lane, dividing it into at least one lane surface according to a preset division method, and determining the lane surface relationships between the at least one lane surface corresponding to each lane; Determining the center line nodes corresponding to the in-plane driving center lines of each lane surface, and determining the node relationships of the center line nodes corresponding to each lane surface according to the lane surface relationships; According to the node relationships, determining the lane relationships between each lane.

3. The method according to claim 2, characterized in that, Before determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes when the lane relationships represent lane intersections, the method further includes: When there is a connection relationship between multiple center line nodes in the node relationships, determining that the lanes corresponding to the lane surfaces where the multiple center line nodes are located are the intersecting lanes, the node where the multiple center line nodes intersect is the intersection node, and the lane relationship is lane intersection.

4. The method according to claim 3, characterized in that, The determining the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes includes: According to the guide area data in the ground semantic data, determining the shortest distance between the guide area and the intersection node; the intersection node is the node among the multiple center line nodes that has a connection relationship with at least two center line nodes; When the shortest distance is less than a preset distance threshold and the traffic direction is the first direction, determining that the node type is divergence; the first direction is the direction in which the intersection node faces the at least two center line nodes; When the shortest distance is less than the preset distance threshold and the traffic direction is the second direction, determining that the node type is confluence; the second direction is the opposite direction of the first direction.

5. The method according to any one of claims 1 to 4, characterized in that, The determining the relative position relationship between the guide area and the intersecting lanes at the intersection node includes: Determining the in-plane driving center line corresponding to the center line node connected to the intersection node as the center line to be analyzed; Extending the center line to be analyzed to obtain an extended center line, and when the extended center line is not a preset special type of center line, using a preset relative position algorithm to determine the relative position relationship based on the center line to be analyzed; Wherein, the relative position relationship includes: the guide area is located in the middle of the intersecting lanes.

6. The method according to claim 1, characterized in that, The guide points include a confluence point, a divergence start point, and a divergence end point; When verifying that the node type is valid according to the relative position relationship, determining the diversion point corresponding to the diversion area according to the node type includes: When the relative position relationship indicates that the diversion area is in the middle of the intersecting lanes, determining that the node type is valid; When the node type is confluence, determining the intersecting node as the confluence point; When the node type is divergence, determining the intersecting node as the divergence starting point, and using a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point; the diversion area control point is the point on the diversion area that is closest to the intersecting node.

7. The method according to claim 1, characterized in that, The method further includes: Based on the side line data of the diversion area, determining the diversion area intersection point of the diversion area, and determining the point on the side line included in the side line data that is closest to the diversion area intersection point as the diversion area control point; When the distance between the diversion area control point and the diversion area intersection point is less than a preset distance, determining the control node type corresponding to the diversion area control point; Based on the control node type, extending the left and right center lines of the obtained diversion area, and when the extended left and right center lines intersect at a point, determining the intersection point after the extension as the intersecting node, and determining the node type of the intersecting node based on the control node type; When the node type is divergence, using a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point.

8. The method according to claim 6 or 7, characterized in that, The divergence end point includes a left divergence end point and a right divergence end point. Using a preset divergence end point algorithm to determine the divergence end point based on the diversion area control point includes: Obtaining the first lane side line adjacent to the left side of the diversion area and the second lane side line adjacent to the right side of the diversion area, and respectively determining the first control point and the second control point on the first lane side line and the second lane side line that are closest to the diversion area control point and the intersecting node; Obtaining the target center lines on both sides of the diversion area, and determining the target center line closest to the first control point among the target center lines as the first target center line, and the target center line closest to the second control point as the second target center line; Determining the projection point of the first control point on the first target center line as the first projection point, and when the distance between the first control point and the first projection point is less than a preset threshold, determining the first projection point as the left divergence end point; Determining the projection point of the second control point on the second target center line as the second projection point, and when the distance between the second control point and the second projection point is less than the preset threshold, determining the second projection point as the right divergence end point.

9. The method according to claim 7, wherein The number of side lines included in the side line data is greater than a first value; based on the side line data of the diversion area, determining the diversion area intersection point of the diversion area includes: According to the side line data, determining at least one side line in the diversion area with the same side line direction and an azimuth angle less than a preset angle threshold, and selecting a target side line from the at least one side line; When the points where the target side line intersects with the side lines in different side line directions satisfy the first preset position condition, determine the intersection points of the diversion areas as the intersection points of the diversion areas.

10. The method according to claim 7 or 9, wherein The number of side lines included in the side line data is the first value. Determining the intersection points of the diversion areas based on the side line data of the diversion areas includes: Obtain the road boundary data of the side line data within a preset range; When the road boundary data indicates the existence of a road boundary, determine the point on the side line included in the side line data that is farthest from the road boundary as the intersection point of the diversion area.

11. The method according to claim 7, wherein After extending the left and right center lines of the obtained diversion area based on the control node type, the method further includes: When the extended left and right center lines do not intersect at a point, obtain the third target center line closest to the left side of the diversion area and the fourth target center line closest to the right side of the diversion area; Determine the projection point of the diversion area control point on the third target center line as the third projection point, and when the third projection point and the third target center line satisfy the second preset position condition, determine the third projection point as the left divergence end point; Determine the projection point of the diversion area control point on the fourth target center line as the fourth projection point, and when the fourth projection point and the fourth target center line satisfy the second preset position condition, determine the fourth projection point as the right divergence end point.

12. An apparatus for determining a diversion point, characterized in that Includes: A determination module, configured to obtain the lane relationship between each lane on the road, and when the lane relationship indicates that the lanes intersect, determine the node type corresponding to the intersection node of the intersecting lanes based on the traffic directions of the intersecting lanes; the node type is confluence or divergence; A verification module, configured to determine the relative position relationship between the diversion area and the intersecting lanes at the intersection node, and when verifying that the node type is valid according to the relative position relationship, determine the diversion point corresponding to the diversion area according to the node type for use in autonomous driving.

13. An equipment for determining a diversion point, characterized in that Includes: A processor, a memory, and a communication bus; The communication bus is used to implement the communication connection between the processor and the memory; The processor is configured to execute the computer program stored in the memory to implement the method for determining the diversion point according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores one or more computer programs, and the one or more computer programs can be executed by one or more processors to implement the method for determining the diversion point according to any one of claims 1 to 11.

Images